Rueff, Allais, et le chômage d’équilibre

The « permanent » unemployment concept of Rueff (1925, 1931) represents a classical form of unemployment due to wage rigidities leading to an excess of real wages compared to their theoretical value corresponding to a competitive equilibrium. Relative to this known aspect of the Rueff’s work, this paper shows that the author considered also a “temporary” unemployment due to an insufficient level of the economic activity, and a « minimum » frictional unemployment prevailing in the normal functioning of any economy. Using empirical data in England during the 1920’s, Rueff suggested that the « permanent » component was the main explanation of the unemployment (this is the socalled « law of Rueff »). We conduct simple econometric tests using Rueff’s data that confirm this conclusion, and we show that releasing the assumption of a constant labor productivity (supposed by Rueff) improves the « law of Rueff ». In line with these results, we show that the theoretical approach proposed by Allais (1943) joins the three types of unemployment pointed out by Rueff, renamed as “chronic”, “conjonctural” and “technological”, respectively. Much later, Allais (1980) proposed an unrecognized straightforward econometrical equation comprising these three types of unemployment to represent the french unemployment over the period 1952-78. We confirm that this equation describes a large part of the evolution of unemployment in England during the 1920’s and in France during the period 1970-2008, although the properties of residuals show that this equation is misspecified. Finally, we suggest that, under some restrictive conditions, the three types of unemployment distinguished by Rueff and Allais can be seen from the perspective of the equilibrium unemployment defined by the imperfect competition WS-PS model (Layard-Nickel-Jackman (1991)), hence allowing to indicate how the Allais’ equation is misspecified, thus highlighting the important scientific advances that have been made since.